一、在这里selenium的作用
（1）模拟的登录。
（2）获取登录成功之后的cookies
代码：
def start_login(self): chrome_options = Options() # 禁止图片加载,禁止推送通知 prefs = { "profile.default_content_setting_values": { "images": 2 }, "profile.default_content_setting_values.notifications": 2 } chrome_options.add_experimental_option("prefs", prefs) if chrome_args().get("headless_flag") == "1": chrome_options.add_argument(chrome_args().get("headless")) chrome_options.add_argument(chrome_args().get("nogpu")) chrome_options.add_argument(chrome_args().get("noinfobars")) chrome_options.add_argument(chrome_args().get("max_windows")) chrome_options.add_argument(self.Proxy_server) driver = webdriver.Chrome(chrome_options=chrome_options) try: get_logger().info("start login.....") try: # login info self.login_name = self.loginInfo.get("login_id") self.password = aes_cbc_decrypt(self.loginInfo.get("login_pwd")) except Exception: get_logger().error("cant get login info,here are detals".format(traceback.format_exc())) wait = WebDriverWait(driver, 30) print(u"start login in") driver.get(self.login_url) try: login_id = wait.until( eccd.presence_of_element_located( (By.XPATH, self.Id_xpath)) ) login_id.send_keys(self.login_name) login_id.send_keys(Keys.ENTER) password = wait.until( eccd.presence_of_element_located( (By.XPATH, self.pwd_xpath)) ) password.send_keys(self.password) submit = wait.until( eccd.presence_of_element_located( (By.XPATH, self.login_btn_xpath)) ) submit.click() # login signal #判断是否显示 右上角是否显示用户名 login_ok = wait.until( eccd.presence_of_element_located( (By.XPATH, self.login_ok_xpath)) ) try: #判断登录成功 if login_ok: get_logger().info("get user name successful:"+u"{}".format(login_ok.text)) try: get_weibo_info=driver.find_element_by_xpath(self.forward_home_page) except: driver.get(self.first_page) forward_home_page_ok = wait.until( eccd.presence_of_element_located( (By.XPATH, self.forward_home_page)) ) forward_home_page_ok.click() time.sleep(5) self.first_page=driver.current_url get_logger().info("get homepage successful,url is {}".format(driver.current_url)) cookies = driver.get_cookies() # 导出cookie get_logger().info("get cookies") get_fansnum_ok= wait.until( eccd.presence_of_element_located( (By.XPATH, self.follow_xpath)) ) self.get_followfanshome(driver) get_logger().info("get_followfanshome ok") self.home_page_source=driver.page_source driver.close() # 关闭chrome #获取主页的 # 如果需要保存cookies self.write_cookie(cookies) get_logger().info("get cookies,login ok") return cookies else: raise RuntimeError('login failed') except: get_logger().error("login failed") return None except: if driver is not None: driver.close() except Exception as e: if driver is not None: driver.close() get_logger().error("dbs operation error，here are details:{}".format(traceback.format_exc())) |
通过使用cookies = driver.get_cookies() #我们获取了cookie那么我们怎用呢。
二、requests如何使用获取到的cookies
1.创建一个Session对象
| req = requests.Session() # 构建Session |
2.转换上面的cookies对象
for cookie in cookies: req.cookies.set(cookie['name'], cookie['value']) # 转换cookies |
3.开始访问该网站的需求页面
data = req.get(url, headers=self.headers).text #获取文本网页 jsondata = req.get(url, headers=self.headers).json() #获取json型网页 |
好了就是这么简单，另外补充一个经常使用的库fake-useragent，可以获取随机的useragent，但是其中谷歌的useragent版本比较低，尤其在访问知乎的时候，会提示浏览器版本低的信息，这一点需要注意。
另外拥有一个headers是对爬虫起码的尊重，所以有个构建好的headers是非常必要的。我就经常用下面的例：
from fake_useragent import UserAgent as UA import random <p>headers = {<br> 'Accept': 'text/html, application/xhtml+xml, image/jxr, <em>/</em>',<br> 'Accept - Encoding': 'gzip, deflate',<br> 'Accept-Language': 'zh-Hans-CN, zh-Hans; q=0.5',<br> 'Connection': 'Keep-Alive',<br> 'User-Agent': UA().random #获取随机的useragent<br> }</p> |
本文内容不用于商业目的，如涉及知识产权问题，请权利人联系51Testing小编(021-64471599-8017)，我们将立即处理。
